a trading strategy based on Elliott Wave Theory - page 120

 
<br / translate="no"> grasn,
You correctly noted that this thread was created by me (back on February 21).
I'm not suggesting anything to you personally, much less rewriting anything.
This is a public forum, I agree, everyone has a say in the matter.
Are you dissatisfied with something?



Not at all. Everything is just fine, at least for me. :о)))
 
Guys, Alex Niroba is back ! What a joy!
Too bad he never learned to be friendly during his absence. He's still trying to satisfy his ego by puffing up his cheeks. But that's all right. Everyone has to go through their own experiences in life. It's all about awareness.

Hear that, Alex? If you are AWARE of yourself and what you are doing in life, you will have much less wasted energy, time and blood, and much less suffering, than if you don't do it. The practice of awareness is a complex and subtle thing. But only through it can you come to your true self.

In fact, Alex has said something right here. This scheme of building channels by TFs (one for each) is the use of fractal structure of the market and that is exactly what Vladislav explained about 40-50 pages ago. Here it is important to choose the right set of f/ffs. I, for example, personally do not agree with the set proposed by Alieh.
 
2 grasn
What "forces" in the forex market do you think could create a potential field?

Hi Sergei !
As far as I can see it, there are two main forces.
1. Fundamentally - counter flows of currencies associated with export-import transactions in the world market.
2. Technically - traders' expectations related to news flow.

I have answered the question, however, I consider this answer to mean nothing. We can start to dig into the nature of market phenomena in order to make a model that reflects these phenomena. It is a complicated, long and probably pointless way to go.

The market reacts quickly or very quickly to any event. Even if such a model is created, it would require a stream of raw data to work. The market receives this data every day continuously. So the model would have to compete with the market on responsiveness. Is it really necessary?

It seems better to use the market itself as a source of raw data, from this data (already homogeneous in nature, unlike economic, political and other news) to reconstruct some universal, abstract market characteristics (for example, potential energy, volatility, etc.) and then use these characteristics to predict future events.

That's exactly what Vladislav's model does.
One should only consciously approach the construction of these characteristics.
 
I think it is better to use the market itself as a source of raw data, from this data (already homogeneous in nature, unlike economic, political and other news) restore some universal, abstract market characteristics (such as potential energy, volatility, etc.), and then use these characteristics to predict future events.


It makes sense, but all these concepts are unfortunately meaningless without a model, a model of any kind, but there has to be one.
 
Yurixx:
It's just rather important to choose the right set of t/fs.

Are you referring to non-standard timeframes?

I don't agree about one channel yet. Although, if you select the correct timeframes, you may limit yourself to one channel per timeframe. Well, if the timeframe is not precisely selected, it seems that we must see the structures of different timeframes in it simultaneously.
 
<br / translate="no"> Which is exactly what Vladislav's model does.
You just have to be conscious of the construction of these characteristics.


I am currently working on selecting such data for searching for potential energy. I already have several variants.


This scheme of building channels by t/f (one for each) is the use of the fractal market structure


Right, but even with this scheme not everything is clear-cut in the choice of one channel and in the selection of periods.
 
Alex Niroba:
are quite common.
and asymmetrical channels, they clearly do not fit into the first approach.

You're absolutely right! Channels are not only parallel,
they can also be convergent and divergent.

By asymmetrical, I meant channels where the boundaries are at different distances from the 'centre' line.
 
Do you mean non-standard timeframes?

I mean the t/f ratio. For example I want to play on local day trends. Then I select М1, М10 and М100. And if I want to play on medium-term trends, I choose H1, H10 and H100. Again, this is all for example.

Although, if we choose the correct timeframes, we can limit ourselves to one channel per timeframe. And if the timeframe is not selected accurately, it seems that we should see the structures of different timeframes in it simultaneously.

Exactly. You got what I was trying to say exactly right.

2 Jhonny
It makes sense, but unfortunately all these concepts are meaningless without a model.

I, for example, like Vladislav's model. It seems to me that it is structured and simple enough to be implemented in a finite time (:-) and that it has a payoff. It is only necessary to add a meaningful use of potential energy to the idea of channels and the method of their construction and maintenance. And also, instead of what Vladislav was silent about, add some zest of your own production and you will get a good trading program.
 
2 Candid
By asymmetrical I meant channels having their borders at different distances from the "midline".

By the way, Alex is absolutely right about converging and diverging channels.

2 grasn
Right, but even with such a scheme not everything is clear-cut in the choice of one channel and in the selection of periods.

This is the complexity of the whole task. And that is why we need a model. It allows at least at some stages to make an unambiguous choice.
And every time you can't make such a choice, you have to enter a parameter. And then we need to find its optimal value. And this is how we get "bad" optimization. And if there are many more parameters, optimization won't help either. Nightmare. :-)
 
Yurixx:
Are you referring to non-standard timeframes?

I mean the t/f ratio.

Do you know how to determine the scale factor? If so, I wouldn't mind a hint of a method idea. :)
Reason: